Been working on a new project.
I needed a astable oscillator that would drive a relay.
Found the NE555 will drive 200mA so seemed ideal.
Got the pcb made and built up.
It has a trimmer to set oscillation speed from 15 seconds to 2 minutes.
Tested it and getting 6 seconds on and 24 seconds off !
Pretty useless.
Investigated the circuit for mistakes shorts etc and seems fine.
The circuit I used was ripped from the interweb and appeared a few times so should be tried and tested.
So scoped the circuit and found the output when driving relay is only going down to 1.5 volts. I use the output through a resistor to charge and discharge the timing cap.
So when output goes high timing is correct but because output low is only going down to 1.5 the delay is shortened greatly.
The spec for the 555 suggested 0.5 volts output for 100mA but my relay is only 30mA so should be well in spec.
But not so.
So either need a better ne555 or a driver transistor for the relay.
Will try a different ne555 first before binning the pcb's.
The pcb's were only £2 from jlcpcb and were bought in with a few other pcb's so no big problem.
I needed a astable oscillator that would drive a relay.
Found the NE555 will drive 200mA so seemed ideal.
Got the pcb made and built up.
It has a trimmer to set oscillation speed from 15 seconds to 2 minutes.
Tested it and getting 6 seconds on and 24 seconds off !
Pretty useless.
Investigated the circuit for mistakes shorts etc and seems fine.
The circuit I used was ripped from the interweb and appeared a few times so should be tried and tested.
So scoped the circuit and found the output when driving relay is only going down to 1.5 volts. I use the output through a resistor to charge and discharge the timing cap.
So when output goes high timing is correct but because output low is only going down to 1.5 the delay is shortened greatly.
The spec for the 555 suggested 0.5 volts output for 100mA but my relay is only 30mA so should be well in spec.
But not so.
So either need a better ne555 or a driver transistor for the relay.
Will try a different ne555 first before binning the pcb's.
The pcb's were only £2 from jlcpcb and were bought in with a few other pcb's so no big problem.
I only use the TI NE variety. I’ve found that all the different flavors of 555 have slightly different behavior at *start up*. One type may initiate the monostable sequence immediately on power up, another may not. Change the stupid thing out and the speaker relay clicks in immediately instead of going thru the delay sequence like the PREVIOUS one did. To get consistent behavior in every placement you need the same one. They say the LM and NE are interchangeable and they are *not*. You can make them do the same thing - with some circuit fiddling to get the reset and control voltage in the right sequence on first power up. I don’t know which one the CMOS emulates. Or if there are different varieties of those too. Probably.
Anything beyond 10 seconds is digital territory. There are lots of Arduino etc chips for cheap. Besides chip variations, you have capacitor leakage, an issue because long time constants require large caps, and PCB leakage. Bipolar 555's have a nasty issue with the "reset" input has to be driven < ~0.5V. Invariably I resorted to op-amps or comparators instead of struggling with the quirks of 555s.
Something else to consider: NE555's do not like inductive loads, even a diode-protected relay drawing a current well within the specs.
Many have been bitten, with unexplainable erratic behaviours as the symptom.
In fact, it is preferable to swap the discharge pin and output, when practical.
The discharge transistor is open-collector, and has less influence on other components via the substrate than the output.
The 555 is cheap, convenient and a seemingly ideal jack of all trades, but it does none of them very well.
It is often advantageous to spend a little more time on a dedicated circuit.
Remember that it was initially designed as a showcase for a kind of rudimentary analog, ASIC like process
Many have been bitten, with unexplainable erratic behaviours as the symptom.
In fact, it is preferable to swap the discharge pin and output, when practical.
The discharge transistor is open-collector, and has less influence on other components via the substrate than the output.
The 555 is cheap, convenient and a seemingly ideal jack of all trades, but it does none of them very well.
It is often advantageous to spend a little more time on a dedicated circuit.
Remember that it was initially designed as a showcase for a kind of rudimentary analog, ASIC like process
I was trying to drive a 5 volt relay
So only getting 3.5 volts out was pretty useless.
I must admit in retrospect when I was looking for a circuit see a few did have relay driver transistors added.
I had used the 555 instead of a small PIC micro as it was a little cheaper.
But the 555 that actually worked was more expensive so it looks like the PIC will be going back into the circuit.
So only getting 3.5 volts out was pretty useless.
I must admit in retrospect when I was looking for a circuit see a few did have relay driver transistors added.
I had used the 555 instead of a small PIC micro as it was a little cheaper.
But the 555 that actually worked was more expensive so it looks like the PIC will be going back into the circuit.
I did a 5 second power up reset for equipment installed in NASA JSC with a 96S02 multivibrator (one shot) and a 1 uf mlcc ceramic cap. Pull up 1 megohm. I buffered the digital output with a couple of 7404's sections to drive the wires all over the system of course. No trouble. Much lower sense current on the capacitor than 555 and a schmitt trigger on the input besides. Smoked a 74121,122,123 which I never used.
96S02 still in stock at rochester via digikey. minimum buy 79 of them @ $3.81.
Nowadays I use a 1n7000 nfet and a ceramic cap. Have to use more parts to provide the digital output, of course. Op amp as schmitt trigger + inverter if the supply is higher than +5.
96S02 still in stock at rochester via digikey. minimum buy 79 of them @ $3.81.
Nowadays I use a 1n7000 nfet and a ceramic cap. Have to use more parts to provide the digital output, of course. Op amp as schmitt trigger + inverter if the supply is higher than +5.
Last edited:
If I use a 555 at all I usually gravitate to the TLC555 CMOS 555 from TI - it's fast and has a respectable source and sink current for a CMOS part. If nothing else is available, I'll hold my nose and use the National CMOS version of the part, (LMC555) which is inferior when it comes to source and sink currents. I never have been able to get the reset pin on any of these parts to work as advertised....
With some intelligence applied, you can trick the device into being a rudimentary controller for a flyback-based DC-DC converter. An even more rudimentary implementation can be done using some CMOS Schmitt trigger gate packages, like the CD4093B or the CD40106B
With some intelligence applied, you can trick the device into being a rudimentary controller for a flyback-based DC-DC converter. An even more rudimentary implementation can be done using some CMOS Schmitt trigger gate packages, like the CD4093B or the CD40106B
Take a look at the ST datasheet
VCC = +5 V IO(sink) = 8mA 0.4V max, they don't spec it at higher current at 5V
As was mentioned previously, 555s are not partcularly keen on reactive loads, either capacitive or inductive. At 5V Vcc, the drive capability is also not fantastic, to say the least. This particular application would have been best served with an external drive transistor to decouple the output from the reactive load. Just about any GP transistor would do as long as the relay coil is clamped with a diode.
Just use a driver transistor.
assume a higher current than 30ma
having played around with these chips long time ago.
I found the best ones from radio shack which were
Texas Instrument manufacture.
Supposedly there is a military package.
which is basically to have better operation in extreme temperatures.
not only extreme heat, but extreme cold.
assume a higher current than 30ma
having played around with these chips long time ago.
I found the best ones from radio shack which were
Texas Instrument manufacture.
Supposedly there is a military package.
which is basically to have better operation in extreme temperatures.
not only extreme heat, but extreme cold.
One thing you should have in your parts box is one bin each for “unknown NPN” and “unknown PNP” transistors. When you salvage a board, yank all the to-92’s, do a quick test and file. When you use a 555, just grab a transistor and you don’t even have to spend a nickel. I almost *never* end up driving anything directly with a 555 - either end up using a PNP or an NPN, depending on the ground reference. This type of thing is perfect for wire-OR-ing with other parallel sensors, typical of amp and power supply supervisor circuits. Two or three things may end up driving the same relay.
The “military” package doesn’t really help with the basic problem. Might help if you want to use your circuit underwater or inside a meat locker (the other components might not like it, tho - especially the timing cap).
The “military” package doesn’t really help with the basic problem. Might help if you want to use your circuit underwater or inside a meat locker (the other components might not like it, tho - especially the timing cap).